基于卷积神经网络及遥感反演的生产建设项目识别方法及应用

doi:10.16843/j.sswc.2022161

摘要
图/表
参考文献
相关文章 (15)

全文: HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要由于遥感影像中的地物普遍存在异物同谱、同物异谱等现象,导致面向对象的多尺度和多阈值分割分类等解译方法无法在完全脱离人工干预的情况下使用,因而给解译带来成本高、效率较低等问题。本研究目的是建立一个实用、高效的生产建设项目扰动图斑自动识别模型,提高扰动图斑识别的效率和准确性。河南省作为水利部第一批遥感监管示范省,至今已积累大量生产建设项目水土保持空间数据。为此,笔者选择河南省作为研究区。提出一种基于神经网络及遥感反演模型的生产建设项目自动识别方法。该方法使用由5个卷积层和3个池化Pooling层的AlexNet对扰动图斑进行初步自动解译,并使用归一化植被指数、温度植被干旱指数等遥感反演模型进一步判断图斑是否为生产建设项目。笔者以成像时间为2020年、覆盖河南全省的GF-1、GF-6遥感影像为基础,使用生产建设项目扰动图斑的外包矩形裁剪上述影像获得训练样本,对样本进行标记,并将标记好的训练样本输入到AlexNet模型对其进行训练,完成训练后分别使用单一神经网络模型及神经网络模型+遥感反演模型进行两组测试。结果表明:1) 本方法在降低误识别率和漏识别率的效果方面均比单纯卷积神经网络识别模型优异,误识别率平均降低23.88%,尤其在降低误识别率方面效果明显;2) 本方法的计算时间,与卷积神经网络识别模型没有明显差异。综上,本研究成果可有效降低生产建设项目自动识别的误判率,有助于进一步提升生产建设项目水土保持遥感监管工作的自动化程度,在生产建设项目自动识别应用领域有一定的借鉴意义。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	卢敬德
	伍容容
	吴鹏
	丁立

关键词 ：卷积神经网络, 遥感反演, 遥感监管, 自动识别, 水土保持

Abstract：[Background] Due to the phenomenon of spectral similarities of foreign objects and spectral differences of the same objects in remote sensing images, object-oriented multi-scale and multi-threshold segmentation and classification methods cannot be used without significant manual intervention. This manual intervention results in high costs and low efficiency in the interpretation process. The purpose of this study is to establish a practical and efficient model for automatic identification of disturbed areas in production and construction projects, aiming to improve the efficiency and accuracy of disturbance identification. As one of the first batch of remote sensing monitoring demonstration provinces designated by the Ministry of Water Resources, Henan province has accumulated a significant amount of spatial data on soil and water conservation for production and construction projects. Therefore, this study selects Henan province as the research area. [Methods] The authors proposed an automatic identification method for production and construction projects based on neural networks and remote sensing inversion models. This method utilized AlexNet, which consists of 5 convolutional layers and 3 pooling layers, for initial automatic interpretation of disturbed spots. Additionally, remote sensing inversion models such as normalized difference vegetation index (NDVI) and temperature-vegetation dryness index (TVDI) were employed to further determine whether the identified spots were production and construction projects. Based on GF-1 and GF-6 remote sensing images taken in 2020, covering the entire Henan province, the author used the external rectangle of suspected production and construction project spot to crop the aforementioned images and obtain training samples. The samples were labeled and then input into the AlexNet model for training. After completing the training, two sets of tests were conducted using a single neural network model and a neural network model combined with remote sensing inversion. [Results] 1) This method performs better than a simple convolutional neural network recognition model in reducing both false recognition rate and missed recognition rate, with an average reduction of 23.88% in false recognition rate. It is particularly effective in reducing false recognition rate. 2) This method does not show any significant difference in terms of runtime cost compared to a simple convolutional neural network recognition model. [Conclusions] This achievement can effectively reduce the misclassification rate in the automatic identification of production and construction projects, contributing to the further improvement of the automation level in remote sensing monitoring of soil and water conservation in production and construction projects. It has certain reference value in the application field of automatic identification of production and construction projects.

Key words： convolutional neural network quantitative remote sensing the remote sensing supervision of soil and water conservation automatic identification soil and water conservation

收稿日期: 2022-07-28

PACS:

TP753

基金资助:河南省水利厅2021年水利科技攻关项目“河南省水土保持大数据平台关键技术研究:生产建设项目水土保持监管系统设计与应用”(GG202151)

通讯作者: 伍容容(1987—),女,本科,工程师。主要研究方向:遥感与地理信息技术。E-mail:yoiko@foxmail.com E-mail: yoiko@foxmail.com

作者简介: 卢敬德(1982—),男,硕士,工程师。主要研究方向:计算机科学与技术,遥感与地理信息技术,水利工程。E-mail:ceo@infmatrix.com

引用本文:

卢敬德, 伍容容, 吴鹏, 丁立. 基于卷积神经网络及遥感反演的生产建设项目识别方法及应用[J]. 中国水土保持科学, 2024, 22(5): 199-208.
LU Jingde, WU Rongrong, WU Peng, DING Li. A methodology of identifying production and construction projects by integrating convolutional neural networks with remote sensing inversion model and its application. SSWC, 2024, 22(5): 199-208.

链接本文:

http://journal12.magtechjournal.com/Jweb_stbc/CN/10.16843/j.sswc.2022161 或 http://journal12.magtechjournal.com/Jweb_stbc/CN/Y2024/V22/I5/199

[1]	中华人民共和国水利部办公厅.水利部办公厅关于推进水土保持监管信息化应用工作的通知:办水保〔2019〕198号[A/OL].(2019-09-11)[2022-07-28].https://www.cnscm.org/xxh/ssjd/202209/t20220903_1320168.html. General Office of the Ministry of Water Resources of the People's Republic of China. Notice from the General Office of the Ministry of Water Resources on Promoting the Application of Information Technology in Soil and Water Conservation Supervision: Office of Water Conservation Document No. 198
	[2019] [A/OL]. (2019-09-11)[2022-07-28].https://www.cnscm.org/xxh/ssjd/202209/t20220903_1320168.html.
[2]	梁顺林. 中国定量遥感发展的一些思考[J]. 遥感学报, 2021, 25(9): 1889. LIANG Shunlin. Some thoughts on the development of quantitative remote sensing in China[J]. National Remote Sensing Bulletin, 2021, 25(9): 1889.
[3]	罗志东, 刘二佳. 基于高分遥感的生产建设项目扰动发现技术研究[C]//中国水土保持学会预防监督专业委员会第九次会议暨学术研讨会论文集, 2015: 403. LUO Zhidong, LIU Erjia. Research on disturbance detection technology of production and construction project based on high resolution remote sensing[C]//Proceedings of the ninth Meeting and Academic Symposium of Prevention and Supervision Committee of Chinese Society for Soil and Water Conservation, 2015: 403.
[4]	GOODFELLOW I, BENGIO Y, COURVILLE A, et al. Deep learning (Vol. 1)[M]. Cambridge, UK: MIT Press, 2016: 326.
[5]	GU Jiuxiang, WANG Zhenhua, JASON Kuen, et al. Recent advances in convolutional neural networks[J]. Pattern Recognition, 2018, 77: 354.
[6]	陈超, 齐峰. 卷积神经网络的发展及其在计算机视觉领域中的应用综述[J]. 计算机科学, 2019, 46(3): 63. CHEN Chao, QI Feng. Review on development of convolutional neural network and its application in computer vision[J]. Computer Science, 2019, 46(3): 63.
[7]	刘强, 解加粉, 陈建忠, 等. 基于卷积神经网络的高分辨率遥感影像建筑物场景识别[J]. 测绘通报, 2021(S1): 124. LIU Qiang, XIE Jiafen, CHEN Jianzhong, et al. Building scene recognition based on convolutional neural network using high resolution remote sensing image [J]. Bulletin of Surveying and Mapping, 2021(S1):124.
[8]	KRIZHEVSKY A,SUTSKEVER I,HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Advances in Neural Information Processing Systems, 2012, 25(2): 1.
[9]	党宇, 张继贤, 邓喀中, 等. 基于深度学习AlexNet的遥感影像地表覆盖分类评价研究[J]. 地球信息科学学报, 2017, 19(11): 1530. DANG Yu, ZHANG Jixian, DENG Kazhong, et al. Study on the evaluation of land cover classification using remote sensing images based on AlexNet[J]. Journal of Geo-Information Science, 2017, 19(11): 1530.
[10]	侯霄雄, 许新征, 朱炯, 等. 基于AlexNet和集成分类器的乳腺癌计算机辅助诊断方法[J]. 山东大学学报(工学版), 2019, 49(2): 74. HOU Xiaoxiong, XU Xinzheng, ZHU Jiong, et al. Computer aided diagnosis method for breast cancer based on AlexNet and ensemble classifiers[J]. Journal of Shandong University (Engineering Science), 2019, 49(2): 74.
[11]	杨洁, 张欢. 基于改进型AlexNet的LPI雷达信号识别[J]. 现代电子技术, 2020, 43(5): 57. YANG Jie, ZHANG Huan. LPI radar signal recognition based on improved AlexNet[J]. Modern Electronics Technique, 2020, 43(5): 57.
[12]	缪裕青, 邹巍, 刘同来, 等. 基于参数迁移和卷积循环神经网络的语音情感识别[J]. 计算机工程与应用, 2019, 55(10): 135. MIAO Yuqing, ZOU Wei, LIU Tonglai, et al. Speech emotion recognition model based on parameter transfer and convolutional recurrent neural network[J]. Computer Engineering and Applications, 2019, 55(10): 135.
[13]	李祥鹏, 闵卫东, 韩清,等. 基于深度学习的车牌定位和识别方法[J]. 计算机辅助设计与图形学学报, 2019, 31(6): 979. LI Xiangpeng, MIN Weidong, HAN Qing, et al. License plate location and recognition based on deep learning [J]. Journal of Computer-Aided Design & Computer Graphics, 2019, 31(6): 979.
[14]	卢敬德, 伍容容, 苏德鹏, 等. 一种禽畜养殖舍棚的自动识别方法、装置及存储介质[P]. 中国: ZL202010396298.8, 2021-1-29. LU Jingde, WU Rongrong, SU Depeng, et al. An automatic identification method, a device and a storage medium for shanty farm[P]. China: ZL202010396298.8, 2021-1-29.
[15]	原琪翔, 尚国琲, 张霞. 基于NDVI的2014—2020年郑州市植被覆盖变化特征分析[J]. 河北地质大学学报, 2022, 45(4): 86. YUAN Qixiang, SHANG Guobei, ZHANG Xia. Analysis of characteristics of vegetation cover change in Zhengzhou city from 2014 to 2020 based on NDVI[J]. Journal of Hebei GEO University, 2022, 45(4): 86.
[16]	赵杰鹏, 张显峰, 廖春华, 等. 基于TVDI的大范围干旱区土壤水分遥感反演模型研究[J]. 遥感技术与应用, 2011, 26(6): 742. ZHAO Jiepeng, ZHANG Xianfeng, LIAO Chunhua, et al. TVDI based Soil moisture retrieval from remotely sensed data over large arid areas[J]. Remote Sensing Technology and Application, 2011, 26(6): 742.
[17]	蒋文君. 地表干旱对城市表面热岛的影响研究:以鄱阳湖地区为例[D]. 南昌: 江西师范大学, 2017: 17. JIANG Wenjun. Study on the effect of surface drought on urban surface heat island in Poyang Lake Area[D]. Nanchang: Jiangxi Normal University, 2017: 17.
[18]	齐述华, 王长耀, 牛铮. 利用温度植被旱情指数(TVDI)进行全国旱情监测研究[J]. 遥感学报, 2003(5): 420. QI Shuhua, WANG Changyao, NIU Zheng. Evaluating soil moisture status in China using the temperature/vegetation dryness index(TVDI)[J]. Journal of Remote Sensing, 2003(5): 420.
[19]	吴鹏, 陈国兵. 河南省生产建设项目遥感监管实践[J]. 中国水土保持, 2020(9): 57. WU Peng, CHEN Guobing. Remote sensing supervision of production and construction projects in Henan province [J]. Soil and Water Conservation in China, 2020(9): 57.